Using Data Mining Techniques In Catalytic Cracking Process Modeling – A Comparative Study

Despite notable successes achieved in trying to obtain viable alternatives to the use of oil as a source of energy and raw materials, oil industry remains one of the main pillars of sustainable development of modern society. Continued modernization of the industry by using the latest technologies in all its important links - extraction, transport, refining - led to substantial reduction of its negative impact on the environment. Particularly, refineries benefited from a special technological contribution, which resulted in a significant plus in efficiency and reliability. In order to always maintain these high standards, the control of oil refining chemical processes must be based on simplebut at the same time performant mathematical models. One way to achieve this objective is the use of modern data mining methods. In this paper we propose to compare the performance of three data mining methods in order to be used in catalytic cracking process modeling in a refinery.